The use of generic process models for process transformation

Transcription

1 The use of generic process models for process transformation S J Childe, P A Smart, A M Weaver University of Plymouth Plymouth, PL4 8AA, U.K., Tel , Fax , [email protected] Abstract The paper extends the work presented previously on the application of generic process models in Business Process Re-engineering (Childe et al 1995). The construction of business process models is contextualised by a five stage composite methodology for BPR. Following an analysis of various process architectures the potential of generic process models is presented. The development of such a model is described. The application of this model in a small manufacturing company is discussed and the areas in which the model proved to be useful are shown. The paper challenges traditional sequential methodologies and suggests a nonsequential orientation and the systemic consideration of six fundamental issues. The paper reports on the current thinking in the area of delivery mechanism development for BPR methodologies. This thinking focuses predominantly on learning and the transfer of relevant expertise to practitioners. Keywords Business Process Re-engineering, methodology, process modelling, generic model 1. INTRODUCTION Over the past five years, Business Process Re-engineering (BPR) has emerged as a popular approach used by organisations seeking improvements in their business performance. This interest began with Hammer and his seminal article entitled Re-engineering work: Don t Automate, Obliterate (Hammer 1990). Surveys by Highams Systems Services Group (Skinner & Pearson 1993) and Business Intelligence (Harvey 1994) report that 65% and 77% respectively of respondents were carrying out or considering Business Process Re-engineering activities. In their survey, Highams Systems Services Group Ltd found that the respondents to the survey were carrying out BPR projects for a number of reasons. These included: the need for continuous improvement increased customer expectations increased competition

2 changing market needs This survey revealed that many companies believe there are a variety of possible benefits to be gained from organising around business processes. However, it also identified that there are many risks associated with BPR projects. It is apparent that no single BPR methodology of those in use by practitioners has been identified as more successful than the others. BPR's rise to prominence as a powerful means of transforming business performance has led to the emergence of a number of gurus in the field, the most notable of whom are Hammer (1990), Harrington (1992) and Davenport (1993). Hammer's intervention strategy, the neutron bomb approach to business improvement (Hammer & Champy 1993) We'll leave the walls standing and we'll nuke everything on the inside, represents one extreme of a wide spectrum of opinion regarding the most appropriate BPR strategies for firms to adopt. Hammer states, for example, that firms can only hope to achieve radical performance improvements using Business Process Re-engineering methods which strive to break away from the old rules about how we organise and conduct business. He states that re-engineering cannot be accomplished in small or cautious steps but must be viewed as an all-or-nothing proposition. Davenport shares this view but concedes that, in practice, most firms will need to combine incremental and radical improvement activities in an ongoing quality programme. He states that before striving for process innovation, a firm will ideally (though not necessarily) attempt to stabilise a process and begin continuous improvement (Davenport 1993). At the other end of the spectrum, Harrington (1992) takes a more incrementalist view. He prefers to use the term Business Process Improvement which he defines as a systematic methodology developed to help an organisation make significant advances in the way in which its business processes operate. The dramatic rise in interest in BPR over the past few years has resulted in a plethora of information, opinions and perspectives on all aspects of BPR. 2. METHODOLOGY FOR BPR The synthesis of many approaches that companies and consultants use for BPR has resulted in the identification of a five stage composite methodology (Maull et al 1995a). phase 1 phase 2 phase 3 phase 4 phase 5 Identify or create corporate, manufacturing and IT strategies Identify key process(es) and performance measures Analyse existing process(es) Redesign and implement process(es) Monitor and continuously improve new process(es)

3 From the experiences of practitioners in the UK it is evident that the majority of effort is expended in phases two and three of the methodology. In the main, these phases require the identification of process boundaries, and the construction of as-is process models highlighting current business situations. 3. PROCESS MODELLING Process models are used to facilitate an organisation s understanding of how it currently operates and what it actually does. To investigate the way in which business processes may be constructed it is necessary to define what a process is. There are many varying definitions (Davenport 1993, Harrington 1991) of a what a business process is, but they generally fail to provide the detail required for identifying process characteristics. 3.1 Process Characteristics Our research into process characteristics has been heavily influenced by the work of Churchman (1971) and Checkland (1981) in the domain of Systems Theory. In our view the concept of a business process within BPR is synonymous with the concept of a system (Weaver et al 1994). A business process is not the same as a function. The people and operations we include in a single business process may come from more than one traditional functional group. A business process can be identified by the business object which flows through it (Childe et al 1994). From this perspective of a business process, it is possible to treat a business process as a purposeful system. In this system (business process) the route to improving its efficiency and effectiveness is the way in which the human resource utilises the physical resources available (Checkland 1982). Therefore these systems or business processes represent Human Activity Systems (HAS). A HAS represents a set of human activities, linked together so that the whole constitutes purposeful activity (Checkland 1981). Systems theory offers various concepts which can be useful for the construction of process models. A useful approach is to consider a system in terms of its parameters (Churchman 1971) namely the: Source; Input; Transformation; Process; Output; Receiver; Feedback. When applying these parameters, analysts may become aware that they are making assumptions about a number of systems concepts, namely: Objectives/purpose; Weltanschauung (Worldview); Boundaries; Resources; Management.

4 Typical Business Process Re-engineering approaches make a number of assumptions about these concepts which are consistent with a hard systems view. Hard systems approaches assume a common Weltanschauung regarding where are we now, a defined set of objectives where do we want to go and, therefore, a means for identifying the action space for change. The weakness in this approach is that business processes represent Human Activity Systems which have a multiplicity of Objectives and Weltanschauungs (Checkland 1981). Consequently the definition of system boundaries, resources and management is highly problematic. 3.2 Framework for analysis A useful framework which may be employed to consider the characteristics of a process are: The development of consensus regarding Weltanschauungs and objectives. Techniques must provide an easy to use/common language for communication and debate. The success of the technique depends upon the simplicity of the syntax and semantics employed by the technique. Consensus needs to be established regarding system boundary and purpose (Root Definition). In addition, a coherent set of modelling rules is required to facilitate the modelling process itself; The description of a process in terms of systems parameters. The identification of a systems hierarchy. The idea of hierarchy is based on a systems concept in which systems are composed of sub-systems (and processes of sub-processes). Harrington (1991) has proposed a hierarchy of macroprocesses, sub-processes, activities and tasks each of which represents further hierarchical detailing of the process. Techniques must facilitate the identification of sub-processes, activities and tasks within the whole process, so that each element of the system is seen in its proper context relative to the other elements, and the analyst can identify the impact of any change upon the whole business system. This framework provides the basis for the identification of appropriate process modelling and analysis techniques for a company. 3.3 Representing a business process using a model The first stage of the analysis of a business process is concerned with constructing a model of the business process. It usually involves the change team modelling the existing or as-is business process. Each model will relate to a particular Weltanschauung of the business process. The creation of an as-is business process model allows the change team to establish common ground and develop a consensus viewpoint. As the change team participates in modelling the as-is business process, they begin to understand what is wrong with it and let go of their attachment to it.

5 The benefits of developing a business process model do not necessarily result from having a complete or totally accurate model. The benefits are more likely to result from the change team communicating their understanding. Successful communication of each perspective will enable the creation of a model that represents a consensus. One of the greatest problems facing most organisations is not deciding what to do differently, but developing the critical momentum required to change existing behaviour. The act of creating a business process model that represents a consensus view can help to develop that critical momentum. It is therefore essential to focus on the modelling of the business process prior to its analysis. We have established that the creation of a business process model is an important activity in the overall BPR project. However, the construction of process models is a resource-intensive activity. This problem may be overcome by utilising a set of generic models which may be adapted to fit specific company scenarios. 4. GENERIC MODELS One of the initial activities of a BPR methodology is the identification of core business processes. This was briefly described in Section 2. In identifying the core processes the participants in the BPR project are defining boundaries within their organisation from a process perspective. 4.1 An architecture of Business Processes The notion of architecture and hierarchy of business processes is discussed by Maull et al (1995b). The provision of a hierarchical architecture both facilitates and simplifies the development of a process model. If the organisation is regarded as a single process, the representation of the whole process may be described as level 0, the most abstract and general level. At the next level in the hierarchy greater process detail is presented. Reported cases identify between 10 and 20 business processes at this level. Business Intelligence's report (Harvey 1994) provides a table of the core process taxonomies of a number of consultants ranging from 7 to 20 core processes. A number of leading organisations have identified between processes at level 1 including BT with 15 processes, Xerox with 11 processes and IBM with 10 processes. Some authors describe a level 1 of between 3 and 5 processes, with between 10 and 20 processes at level 2. Parnaby (1993) describes 3 processes at level 1 and 16 at level 2 at Lucas. Other similar hierarchies are described by Pagoda (1993), Harrington (1984), Veasey (1994). Porter (1985) identifies primary and support activities in his value chain concept. The primary activities are essentially value adding from an external customer perspective, and support activities are those that enable the primary activities to function.

6 The CIM-OSA standard provides a recognised framework around which to group the processes identified by organisations. The first two types of processes are the Operate processes that add value and the Support processes that enable the Operate processes to function. Operate processes are viewed as those which are directly related to satisfying the requirements of the external customer, for example the logistics supply chain from order to delivery. A detailed description of the Operate dimension of the framework is discussed in Maull et al (1995b). The Support processes include the financial, personnel, facilities management and information systems provision activities. The third type of process, the Manage processes, are the processes that develop a set of business objectives, a business strategy and manage the overall behaviour of the organisation. The grouping of the processes under Manage, Operate and Support provides a framework for the classification of process types. 4.2 An IDEF 0 model of standard business processes The development and validation of a model of four standard business processes that may be classified as Operate processes is described by Maull et al (1995). In this document the model and guidelines for its use are included as an appendix. The purpose of developing and validating a set of standard business models was to help individuals in a manufacturing company identify and understand their key business processes. The IDEF 0 model created represents four Operate processes. The model shows the interactions between each of the processes and external customers, suppliers and other parts of the organisation that are outside the boundaries of the high level processes. IDEF 0 allowed a model of each process to be developed separately and then combined into a single model of the Operate processes. The model includes a set of IDEF 0 diagrams and a glossary of terms. The completed model represents an integrated set of over 180 different activities and over 250 different flows. The model extends to six levels of decomposition. It does not show who or what performs any of the activities represented. For a specific organisation the identification of who or what performs an activity could be one of the first stages of adapting the standard model to represent their own business processes. The model may include activities that would be removed or changed in the vast majority of reengineering projects, for example inspection activities. The only criterion for including activities and flows in the model was that the activity or flow under consideration would normally form part of the business process in a typical manufacturing organisation. The model is not intended to represent best practice. 4.3 Validation of the model The model has been validated in two stages. The first stage was to ensure that the model represented a consensus view of the Operate processes in a typical manufacturing company. The second stage was to ensure that the model could be used to help a manufacturing company gain an understanding of their business processes.

7 4.4 Assessment of the model by practitioners The first stage of the validation involved the assessment of the model by practitioners. Over 85 copies of the model were distributed to practitioners who had responded to a series of requests for reviewers. Of the 85 practitioners who received the model for review, 29 provided feedback. The reviewers were employed in a variety of areas including manufacturing organisations, consultancy and academia and many were outside the UK. The feedback provided by the reviewers was positive and many issues raised were acted upon to improve the model. The main issues included: Alterations that needed to be made to the model; Complexity of the model; Review of some of the terms used to avoid any ambiguity; Explicit statement of the viewpoint from which the model was created; The type of manufacturing companies to which the model was applicable; The use of IDEF 0 as the modelling technique. 4.5 Use of the model in a local manufacturing company The second stage of the validation involved the use of the model by the management of a local manufacturing company to understand their key processes and develop an as-is model of their order fulfilment process. A single briefing on how the model was intended to be used was presented to the IT manager of the company. The IT manager recounted his experiences in using the generic model after the development of an as-is model had been completed. During a series of interviews with key personnel the IT manager used the model to discuss activities carried out by the interviewee, the flows between activities and the context of the activities within an overall order fulfilment process ( Walking through the process). The discussion resulted in a common understanding and view of what was happening within the company as part of the process. The IT manager identified four main areas in which the model had proved useful to the company. These areas were: allocation of responsibility - adding mechanisms to indicate who was responsible for activities initiated discussions about who was and should be responsible for activities and supplying information. taken for granteds - by walking through the process discussions about activities that had not been properly considered before were initiated. ideas for improvement - some activities shown on the model were not carried out by the company. After further discussions it was decided that the company would benefit from carrying out these activities.

8 time saving - the IT manager stated that developing an as-is model using the standard process model took considerably less time than developing an as-is model from a blank sheet of paper. The company did make a number of recommendations concerning improvements to the model. They are currently using the models as a basis for specifying a new CAPM system. The generic models may be seen to facilitate the identification of core processes and the construction of process models. The generic model is not a prescriptive solution, or a representation of best practice. Rather, it is a tool which may be used to help and support manufacturing companies in their BPR programmes. As indicated in section 2, the second phase of a BPR methodology is typically concerned with the identification and modelling of core processes. 5. FOCUS OF BUSINESS PROCESS RE-ENGINEERING Research conducted in four case study companies and the visits to four leading organisations provided some insight into the typical focus of BPR programmes (Maull et al 1995a). Despite the variations in size, complexity and industrial sector, a set of fundamental issues were found. These issues can be described under six main headings; scope of change performance measures information technology human factors business process architecture strategy for BPR In our view each company was having to address these issues at some point in the life cycle of their BPR project. Strict adherence to the set of general methodological phases presented in Section 2 may be in danger of misleading BPR practitioners. A company may go through each phase assiduously yet never consider fully the six issues now identified. This will affect the way in which the methodology is implemented. For example, a process architecture structure aids the identification of processes in Phase Two of the methodology. It also helps the analysis of these processes and acts as a catalyst in Phase Three. Thus the existence of a process architecture will affect the way the BPR project is carried out. Similarly a focus on radical rather than incremental change may be concerned with phase three (re-design), and may not require the analysis of existing processes. Organisations may focus their attention exclusively on incremental product based processes and omit any strategic considerations. Others may focus on developing an infrastructure for change based around flexible working. A Benchmarking project may use a process structure (Childe and Smart 1995) without articulating any methodology. Thus there is a need to challenge the step by step

9 nature of these methodologies. Two essential aspects of a BPR methodology need to be addressed; 1. A non-sequential orientation. 2. Systemic consideration of the six issues identified. 6. POTENTIAL DELIVERY MECHANISM Research is being undertaken to identify appropriate delivery mechanisms for a BPR methodology. The issue of a non-sequential approach may be accommodated in a computer based tool designed on a hypermedia structure. The functional perspective of hypermedia is that it has a node and link structure which enables the user to explore its content in a non-linear fashion (Knussen et al 1991). Ensuring the systemic consideration of the six issues identified, the education of the members of the change team may need to be addressed. In addition to the non-sequential structure provided by hypermedia, it is recognised that the processes of association and exploration that are vital to learning and maintaining user motivation are also facilitated by a hypermedia structure. Historically, the capabilities of the emerging and evolving technologies have been used as the basis for the design of learning sequences irrespective of any model of learning. Consequently many instructional technologies have yielded disappointing results (Hannafin and Philips 1987). It is therefore important to consider research in the field of education, specifically models of instructional design, prior to the design and development of learning mechanisms for BPR. 7. CONCLUSION This paper has addressed the development and use of generic process models for manufacturing enterprises. Following rigorous validation of four process models classified as Operate processes the models were applied in an industrial company. The models were found to be useful by this company in the areas of; the allocation of responsibility, addressing taken for granted assumptions, the provision of ideas for improvement, and the time saved in constructing process models. The resource intensive nature of the modelling phase of BPR was indicated and contextualised by a composite five stage methodology. Further research has challenged this methodological approach and indicated six fundamental issues which should be addressed by a BPR project. Two essential areas of BPR methodologies are identified as major research issues. These issues include the non-sequential orientation of methodological approaches and the need to systemically consider the six issues identified. Research on potential delivery mechanisms for a BPR methodology has focused on hypermedia environments which provide the transfer of expertise. A requirement of the design of these mechanisms needs to address models of human learning and the research which has been undertaken in the field of instructional design.

10 References Checkland P B, (1981), Systems thinking, systems practice, J Wiley and Sons, New York Checkland P, (1982), Systems concepts relevant to the problem of integrated production Systems in Wilson B (ed) Efficiency of Manufacturing Systems, NATO Series. Childe, S J, Maull, R S, Bennett, J, (1994), Frameworks for understanding business process reengineering, International Journal of Operations and Production Management, Vol.14, No.12, Childe S J, Weaver, A M, Maull, R S, Smart, P A, Bennett, J (1995), The application of generic process models in business process re-engineering, in Browne, J., O Sullivan, D.(eds), Re-engineering the Enterprise - Proceedings of the IFIP TC5/WG5.7 Working Conference on Re-engineering the Enterprise, Galway, Ireland., Childe S J, Smart, P A, (1995), The use of process modelling in benchmarking, in Rolstadas, A., (ed), Benchmarking - Theory and Practice - Proceedings of the IFIP WG5.7 Workshop on Benchmarking - Theory and Practice, Trondheim, Norway, June (1994), Churchman C W, (1971), The design of inquiring systems, Basic Books, NY Davenport T H, (1993), Process Innovation, Harvard Business School Press, USA Hammer M, (1990), Reengineering Work: Don't Automate, Obliterate, Harvard Business Review, July-August Hammer M, Champy J, (1993), Re-engineering the corporation, HarperCollins, New York Hannafin, M J, Phillips, T. L., (1987), Perspectives in the design of interactive video, Journal of Research and Development in Education, Vol 21, No. 1, Fall, Harrington J, (1984), Understanding the manufacturing process, Marcel Dekker Inc, New York Harrington H J, (1992), Business Process Improvement, McGraw-Hill Harvey D, (1994), Re-engineering: The Critical Success Factors, Business Intelligence Ltd, UK Knussen, C, Tanner, R G, Kibby, M R, (1991), An approach to the evaluation of hypermedia, Computers and Education, 17, 1, Maull, R S, Weaver, A M, Childe, S J, Smart, P A, Bennett, J., (1995a), Current issues in business process re-engineering, International Journal of Operations and Production Management, Vol.15, No.11, Maull, R S, Childe, S J, Bennett, J, Weaver, A M, Smart, P A,(1995b), Different types of manufacturing processes and IDEF 0 models describing standard business processes, working paper WP/GR/J , University of Plymouth, August. Pagoda Associates Ltd, (1993), Pagoda Process Management Report, London Parnaby J, (1993), Business Process Systems Engineering, Lucas Industries plc, November Porter M E, (1985), Competitive Advantage: Creating and Sustaining Superior Performance, The Free Press, NY Skinner, C, Pearson, J R W, (1993), Business Process Re-engineering in the UK Financial Services Industry, Highams Systems Services Group Ltd. Veasey P W, (1994), Managing a programme of business re-engineering projects in a diversified business, Long Range Planning, Vol 27 No 5 Weaver A, Maull R, Childe S, Bennett J., (1994), A Soft Systems Perspective of Business Process Re-engineering, International Conference on Systems Engineering, Coventry, Sept